What I meant Jon is that because these lights all have automatic temp regulation, I think a smaller light with less thermal mass would reach upper temp limit faster and hence invokes step-down faster. The bigger light should have more prolonged brightness, especially at Turbo starting level.
Above is true in my admittedly limited experience of 219b 4500k in Convoy S21D vs Emisar D4V2. Granted there might be other factors (drivers, etc.) that affect brightness, but with both thermal limits set to 55, the Emisar dims not just to a lower level, but also does it sooner. (I still love my Emisar though because I like its tiny size and it’s a very pretty light.)
The Opple Pro/Series 3 is here! A mere 10 days from China to my house in LA. It’s brand new and working as expected. Again, if you’d like to get one, just make sure it’s the item on the right side of the 2 pictures on the web site https://www.aliexpress.com/item/3256803116495145.html?spm=a2g0o.order_list.0.0.4f2f18026hLcUc, where under “Color” it will say Pro (please don’t ask me why it’s under “Color”). The other way to tell is of course the Pro is more expensive, ~$40.
The Nichia LEDs’ numbers measured with my Opple Pro are fairly consistent with known specs and findings of other posters.
For example the S21D with 219b sw45k - some preliminary trials’ graph & numbers below. The test may not add to what is already known, but still it is fun to confirm for yourself. OCD brains tend to have doubts and if that’s the case this Opple is a godsend .
FWIW this is Simon’s numbers (I believe) measured with the $600 Hopoocolor meter:
CRI for 219b sw45k Ra 96, R9 97, R12 79, DUV –0.0095
219b 4500k My numbers: CCT 4486 Ra 95.8, x 0.3582 y 0.3446 Duv –0.0087
agree, the 4500K 519a runs low, I just tested one of mine at 4200K, duv –0.0005
the nominal CCT that an LED is binned at is not exact, LEDs vary, in this case by 7%.
imo metered values vary due how bright the light is when tested, in addition to differences in the LEDs from one to the other, plus differences between the instruments used to do the test.
Even LEDs from the same roll, side by side neighbors, can have these variations.
Next up is my measurement of Convoy S21D with Nichia 519a 4500k shows 2 interesting findings:
CCT is ~4100k, not 4500k.
Duv is negative, versus the positive number that Simon got using his Hopoocolor meter.
I was surprised by the negative number, considering the beam color looks more yellow/slight greenish than magenta if compared back to back with 219b, so I posed the question on the lightmaster thread Opple Light Master III (G3) discussion thread (Cheap device for measuring Lux, CCT + CRI) - #392 by cannga and found other people seem to be getting negative numbers also. No big deal since variations are expected.
Simon’s specs for 519a 4500k: Ra 93.6, R9 82, R12 80, Duv 0.00091
(Not bad at all. Check out R9 ——> 82. In other words this is also a R9080 led.)
My 519a 4500k measurements with Opple Lightmaster Pro (trial runs), in Convoy S21D with Sofirn Clear Optic (the OEM frosted optic’s numbers didn’t change much):
100% Brightness: Ra 97.3, Duv –0.0037
20% Brightness: Ra 97.6, Duv –0.0018, range: –0.0020 to –0.0030
Agreed. I should emphasize the only time I see any greenish tint in 519a 4500k is during back to back comparison with 219b 4500k or 3500k. By itself, it’s a beautiful golden light.
Jon, very good points to keep in mind (comparison more valid when not compared across different test devices, brightness, LED batch, etc.). That way we could stay relaxed and have fun while learning something.
I’m sure you’ve noticed there are 3 zero’s in front of Simon’s value. So his number hugs the BBL very close, maybe not a “huge” difference from my 20% brightness value. ie Simon’s ~ .001 (round up from the seemingly large 0.0009) versus my .002, and your.000.